[GH-ISSUE #16] [Bug]: Not able to see any log in PENTAGI UI #8

New Issue

2026-06-06T22:07:54-04:00

yindo commented

2026-06-06 22:07:54 -04:00

Originally created by @Girrajjangid on GitHub (Mar 23, 2025).
Original GitHub issue: https://github.com/vxcontrol/pentagi/issues/16

Originally assigned to: @asdek on GitHub.

Affected Component

Core Services (Frontend UI/Backend API)

Describe the bug

as mentioned in readme.md for instructions to start the application I followed all of them correctly.
After adding OPENAI_KEY. I run this command.

curl -O https://raw.githubusercontent.com/vxcontrol/pentagi/master/docker-compose.yml
docker compose up -d

I also added LANGFUSE KEYS and I am able to get Trace on langfuse Web.

After sending first instruction. I do not see anything on PenAGI its of kind stuck. But I can see 3 logs on langfuse web. Only 3.
Any possible reason ?

I am using windows 11 with docker desktop

Steps to Reproduce

Open the https://localhost:8443/chat/new
enter the instruction.
Then it will not show anything. the entered text stuck inside text editor. nothing is happening.

System Configuration

Using WIndows 11.
RAM 16GB
DISK 1TB

Logs and Artifacts

No response

Screenshots or Recordings

No response

Verification

I have checked that this issue hasn't been already reported
I have provided all relevant configuration files (with sensitive data removed)
I have included relevant logs and error messages
I am running the latest version of PentAGI

Originally created by @Girrajjangid on GitHub (Mar 23, 2025). Original GitHub issue: https://github.com/vxcontrol/pentagi/issues/16 Originally assigned to: @asdek on GitHub. ### Affected Component Core Services (Frontend UI/Backend API) ### Describe the bug as mentioned in readme.md for instructions to start the application I followed all of them correctly. After adding OPENAI_KEY. I run this command. ``` curl -O https://raw.githubusercontent.com/vxcontrol/pentagi/master/docker-compose.yml docker compose up -d ``` I also added LANGFUSE KEYS and I am able to get Trace on langfuse Web. After sending first instruction. I do not see anything on PenAGI its of kind stuck. But I can see 3 logs on langfuse web. Only 3. Any possible reason ? I am using windows 11 with docker desktop ### Steps to Reproduce 1. Open the https://localhost:8443/chat/new 2. enter the instruction. 3. Then it will not show anything. the entered text stuck inside text editor. nothing is happening. ### System Configuration Using WIndows 11. RAM 16GB DISK 1TB ### Logs and Artifacts _No response_ ### Screenshots or Recordings _No response_ ### Verification - [x] I have checked that this issue hasn't been already reported - [ ] I have provided all relevant configuration files (with sensitive data removed) - [ ] I have included relevant logs and error messages - [x] I am running the latest version of PentAGI

yindo added the bug label 2026-06-06 22:07:54 -04:00

yindo closed this issue

2026-06-06 22:07:54 -04:00

yindo commented

2026-06-06 22:07:55 -04:00

@asdek commented on GitHub (Mar 24, 2025):

Hello @Girrajjangid

This may be related to the fact that it is currently downloading the Docker image needed to run your scenario.

Could you provide the latest 10 log lines from the running Docker container PentAGI to assess what problems are shown in the logs?

I plan to add the current scenario execution steps to the UI, to understand whether the Docker image is currently loading or the agent has moved to the next step of using this Docker image.

@asdek commented on GitHub (Mar 24, 2025): Hello @Girrajjangid This may be related to the fact that it is currently downloading the Docker image needed to run your scenario. Could you provide the latest 10 log lines from the running Docker container PentAGI to assess what problems are shown in the logs? I plan to add the current scenario execution steps to the UI, to understand whether the Docker image is currently loading or the agent has moved to the next step of using this Docker image.

yindo commented

2026-06-06 22:07:55 -04:00

@Girrajjangid commented on GitHub (Mar 24, 2025):

This is the log.

This is Web UI.

This is Langfuse Logs.

@Girrajjangid commented on GitHub (Mar 24, 2025): This is the log. ![Image](https://github.com/user-attachments/assets/0023a708-2286-4857-b24e-b8ac199bb17f) This is Web UI. ![Image](https://github.com/user-attachments/assets/96c28e14-05ec-4064-9054-c69bcbb43845) This is Langfuse Logs. ![Image](https://github.com/user-attachments/assets/d198d52b-5b92-4dad-9707-fb2a67b9a8a2)

yindo commented

2026-06-06 22:07:55 -04:00

@asdek commented on GitHub (Mar 25, 2025):

@Girrajjangid thank you for additional information.

From your first screenshot, I can see that PentAGI is currently downloading the Kali Linux Docker image. This image is over 6 gigabytes in size, and the download process can take quite some time. In an upcoming pull request, I will add functionality to display the image download progress within the product interface so you can see when the download is complete.

In the meantime, I suggest manually downloading the image by running the docker pull command with the image name shown in your screenshot, and wait for it to finish. After the image is downloaded, PentAGI will use the locally cached image from your Docker registry, and should proceed without delay.

@asdek commented on GitHub (Mar 25, 2025): @Girrajjangid thank you for additional information. From your first screenshot, I can see that PentAGI is currently downloading the Kali Linux Docker image. This image is over 6 gigabytes in size, and the download process can take quite some time. In an upcoming pull request, I will add functionality to display the image download progress within the product interface so you can see when the download is complete. In the meantime, I suggest manually downloading the image by running the docker pull command with the image name shown in your screenshot, and wait for it to finish. After the image is downloaded, PentAGI will use the locally cached image from your Docker registry, and should proceed without delay.

yindo commented

2026-06-06 22:07:55 -04:00

@Girrajjangid commented on GitHub (Mar 26, 2025):

Thanks @asdek it worked. Amazing work guys. .
Just one question. How to stop the API call execution. Let say I don't want to send any request to LLM how to stop that.
I tried multiple ways to stop it. But it keep running.

I log out and log in back. But it still sending request to OpenAI
I shutdown docker and up again but it still sending request to OpenAI.
Then I manually change openai key to incorrect then it stop.

So, Is there a way on UI to stop the execution. Because Agents run in infinite loop.

@Girrajjangid commented on GitHub (Mar 26, 2025): Thanks @asdek it worked. Amazing work guys. . Just one question. How to stop the API call execution. Let say I don't want to send any request to LLM how to stop that. I tried multiple ways to stop it. But it keep running. 1. I log out and log in back. But it still sending request to OpenAI 2. I shutdown docker and up again but it still sending request to OpenAI. Then I manually change openai key to incorrect then it stop. So, Is there a way on UI to stop the execution. Because Agents run in infinite loop.

yindo commented

2026-06-06 22:07:55 -04:00

@asdek commented on GitHub (Mar 26, 2025):

@Girrajjangid

To stop the ongoing execution of the agent, you can use the user interface of the application. Here's how:

Click on the three dots menu associated with the Flow you want to stop. This is usually located next to the Flow's name or at the top left corner of the interface.
Select "Stop Flow" from the dropdown menu. This action will halt the agent's execution and prevent it from sending any further requests to the OpenAI API.

Regarding the agent continuing to run after restarting Docker or the application:

Automatic Restart Behavior: The system is designed to automatically restart the Flow and resume the attack simulation after a docker daemon, host or container application restart. This is intentional and considered standard functionality. It's meant to ensure that simulations can continue uninterrupted even if the system experiences a restart.

Thank you for your feedback. Let me know if you need further assistance! 😉

@asdek commented on GitHub (Mar 26, 2025): ![Image](https://github.com/user-attachments/assets/acf9dd94-657d-486e-a935-7ab193eba434) @Girrajjangid To stop the ongoing execution of the agent, you can use the user interface of the application. Here's how: 1. **Click on the three dots menu** associated with the Flow you want to stop. This is usually located next to the Flow's name or at the top left corner of the interface. 2. **Select "Stop Flow"** from the dropdown menu. This action will halt the agent's execution and prevent it from sending any further requests to the OpenAI API. Regarding the agent continuing to run after restarting Docker or the application: - **Automatic Restart Behavior**: The system is designed to automatically restart the Flow and resume the attack simulation after a docker daemon, host or container application restart. This is intentional and considered standard functionality. It's meant to ensure that simulations can continue uninterrupted even if the system experiences a restart. Thank you for your feedback. Let me know if you need further assistance! 😉

yindo commented

2026-06-06 22:07:55 -04:00